Project Summary Actin cytoskeleton dynamics and membrane dynamics are often interconnected and tightly regulated. BAR domain-containing proteins are emerging as a critical linkage between signaling, the cytoskeleton and membranes. The BAR domain is a dimerization, membrane-curvature sensing/inducing module that occurs in modular proteins in association with other domains, including actin cytoskeleton regulatory, auto-inhibitory, and signaling modules. While the study of BAR domain proteins has recently intensified, what is critically lacking is a comprehensive structure-function understanding of the interplay between their membrane-binding, cytoskeleton-regulatory and signaling activities, which is the goal of this proposal. Initially, the focus will be on three proteins: PICK1, IRSp53 and PInB. PICK1 has emerged as a key regulator of AMPA receptor trafficking in neuronal cells, a process linked to synaptic plasticity, learning, and memory. IRSp53 is enriched in synapses, and is implicated in the formation of neuronal spines and cellular protrusions such as lamellipodia and filopodia. PInB had never been characterized, but preliminary studies presented here suggest that it stabilizes the brush border membrane of epithelial cells. IRSp53 and PInB share moderate sequence identity (24%), and will be studied in parallel, because it is anticipated that these two proteins share similar functional mechanisms and binding partners. Aim 1 will test the hypothesis, emerging from preliminary studies, that PICK1 functions as a scaffold linking membrane vesicles and myosin motors for receptor trafficking in neurons. Another hypothesis suggested by the pilot studies that will be tested is that PICK1 is internally auto-inhibited in the resting state, and becomes activated by coordinated interactions of its various domains with receptor tails, membranes and myosin motors. Aim 2 will test the hypothesis that PInB represents a fundamentally new type of BAR domain protein, involved in the formation of planar membrane structures in epithelial cells. The mechanisms of auto-inhibition and activation by Rho-family GTPases of IRSp53 and PInB will be investigated. Binding partners of the SH3 domain of PInB will be identified in cells, and their interactions will be characterized. Extensive preliminary results lay the groundwork for these studies. Nearly all the protein constructs have been expressed and characterized. Full-length PICK1 was crystallized with bound Ca2+ and the GluR2 AMPA receptor tail. The structure of the BAR domain of PInB is nearly finished. Collaborative cellular studies on PICK1 and PInB have already produced important results and, more importantly, the feedback between the cellular and structural/biophysical studies is beginning to generate new hypotheses.